Synchronous Ethernet and IEEE 1588 in Telecoms E-Book

Jean-Loup Ferrant

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 568

Veröffentlichungsjahr: 2013

Beschreibung

This book addresses the multiple technical aspects of the distribution of synchronization in new generation telecommunication networks, focusing in particular on synchronous Ethernet and IEEE1588 technologies. Many packet network engineers struggle with understanding the challenges that precise synchronization distribution can impose on networks. The usual "why", "when" and particularly "how" can cause problems for many engineers. In parallel to this, some other markets have identical synchronization requirements, but with their own design requirements, generating further questions. This book attempts to respond to the different questions by providing background technical information. Invaluable information on state of-the-art packet network synchronization and timing architectures is provided, as well as an unbiased view on the synchronization technologies that have been internationally standardized over recent years, with the aim of providing the average reader (who is not skilled in the art) with a better understanding of this topic. The book focuses specifically on synchronous Ethernet and IEEE 1588 PTP-based technologies, both key developments in the world of synchronization over the last 10 years. The authors address the needs of engineers and technical managers who are struggling with the subject of synchronization and provide an engineering reference for those that need to consider synchronization in NGN. The market applications that are driving the development of packet network synchronization and timing architectures are also discussed. This book provides a wide audience with everything they need to know when researching, implementing, buying and deploying packet synchronization architectures in telecommunication networks.

Leseprobe

Table of Contents

Foreword

Abbreviations and Acronyms

Acknowledgments

Introduction

I.1. The importance of synchronization in future telecommunications networks

I.2. Purpose of this book

I.3. Differences between frequency and phase/time

I.4. From traditional TDM synchronization to new mobile applications

I.5. Structure of the book

I.6. Standardization

I.7. Bibliography

Chapter 1 Network Evolutions, Applications and Their Synchronization Requirements

1.1. Introduction

1.2. Evolution from plesiochronous digital hierarchy to optical transport networks

1.3. Migration and evolution in the next-generation networks: from time division multiplexing to packet networks

1.4. Mobile networks and mobile backhaul

1.5. Synchronization requirements in other applications

1.6. The need to define new synchronization technologies

1.7. Bibliography

Chapter 2 Synchronization Technologies

2.1. Fundamental aspects related to network synchronization

2.2. Timing transport via the physical layer

2.3. Packet timing

2.4. IEEE 1588 and its Precision Time Protocol

2.5. The concept of “profiles”

2.6. Other packet-based protocols

2.7. GNSS and other radio clock sources

2.8. Summary

2.9. Bibliography

Chapter 3 Synchronization Network Architectures in Packet Networks

3.1. The network synchronization layer

3.2. Functional modeling

3.3. Frequency synchronization topologies and redundancy schemes using SyncE

3.4. The IEEE 1588 standard and its applicability in telecommunication networks

3.5. Frequency synchronization topologies and redundancy schemes using IEEE 1588

3.6. Time synchronization topologies and redundancy schemes

3.7. Bibliography

Chapter 4 Synchronization Design and Deployments

4.1. High-level principles

4.2. MAKE or BUY network synchronization strategies

4.3. Deployment of timing solutions for frequency synchronization needs

4.4. Deployment of timing solutions for accurate phase/time synchronization needs

4.5. Bibliography

Chapter 5 Management and Monitoring of Synchronization Networks

5.1. Introduction

5.2. Network management systems and the telecommunications management network (TMN)

5.3. Synchronization Network management: the synchronization plan and protection

5.4. Provisioning and setup: manual versus automatic

5.5. Monitoring functions

5.6. Management issues in wireless backhaul

5.7. Network OS integration: M.3000 versus SNMP

5.8. Bibliography

Chapter 6 Security Aspects Impacting Synchronization

6.1. Security and synchronization

6.2. Security of the timing source

6.3. Security of synchronization distribution

6.4. Synchronization risk management

6.5. Bibliography

Chapter 7 Test and Measurement Aspects of Packet Synchronization Networks

7.1. Introduction

7.2. Traditional metrics

7.3. Equipment configuration

7.4. Reference signals, cables and connectors

7.5. Testing Synchronous Ethernet

7.6. Testing the IEEE 1588 end-to-end telecom profile

7.7. Bibliography

Appendix 1 Standards in Telecom Packet Networks Using Synchronous Ethernet and/or IEEE 1588

A1.1. Introduction

A1.2. General content of ITU-T standards

A1.3. Summary of standards

A1.4. Bibliography

Appendix 2 Jitter Estimation by Statistical Study (JESS) Metric Definition

A2.1. Mathematical definition of JESS

A2.2. Mathematical definition of JESS-w

Permissions and Credits

Biography

Index

First published 2013 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address:

ISTE Ltd27-37 St George’s RoadLondon SW19 4EUUK

www.iste.co.uk

John Wiley & Sons, Inc.111 River StreetHoboken, NJ 07030USA

www.wiley.com

© ISTE Ltd 2013

The rights of Jean-Loup Ferrant, Mike Gilson, Sébastien Jobert, Michael Mayer, Laurent Montini, Michel Ouellette, Silvana Rodrigues, Stefano Ruffini to be identified as the authors of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988.

Library of Congress Control Number: 2013933689

British Library Cataloguing-in-Publication Data

A CIP record for this book is available from the British Library

ISBN: 978-1-84821-443-9

Printed and bound in Great Britain by CPI Group (UK) Ltd., Croydon, Surrey CR0 4YY

Foreword

Synchronization is the bedrock of the telecommunication highways. Much like the pavement of a highway, synchronization is often taken for granted, since it is effectively invisible when it is working. But it is an enabler of many aspects of the transfer of voice and data. Some services directly require synchronization. More specifically, synchronization can optimize the use of a given bandwidth, increasing available throughput in a fixed band.

Time-and-frequency issues are rich and complex enough that they form their own discipline. Unfortunately, there are few places of study that have this unique focus as an institute unto itself. Hence, the importance of this book. There is much confusion about principles of time and frequency, in part because our human experience of time is so intriguing. Time is a major focus in human culture, in art, philosophy, and song. Yet time in science, engineering and metrology is a different thing.

Scientific time and frequency start with a clock, a device that realizes a theoretical principle in a physical way. The underlying principle in any clock is always a law of physics that predicts circumstances in which the states of a system will repeat at a constant rate. A clock physically realizes this theoretical principle and produces this rate, or frequency, with some level of accuracy and stability. The underlying principle is a theory of physics that forces the theoretical rate of the clock to be constant by definition. Time from a clock comes by counting the states as they repeat themselves, just as counting days produces the calendar. Hence, a clock fundamentally produces a frequency, with time optionally produced by using a counter. The standard frequency is a physical quantity, equivalent to an energy. The time standard, however, is a man-made artifact. A standard frequency signal can be produced by a single device, a cesium standard. To get standard time from a clock, as opposed to time intervals, the clock must first be set, or synchronized, against a reference. Then, since the clock is at best a frequency device with some white noise on the signal, any two clocks will wander off from each other in time without bounds if they are not re-synchronized periodically, whereas the best clocks may be bounded in their native frequency differences.

So where does the time reference come from? Metrologically, the practical time standard is a weighted average of clock times from all over the world. This is produced by the International Bureau of Weights and Measures (Bureau International des Poids et Mesures – BIPM) in the forms of International Atomic Time (TAI) and Universal Coordinated Time (UTC). These time scales are produced only after the fact. Any real-time time signal can be only a prediction of what the correct time will be when it is defined later.

The result of these facts is that a system that requires only frequency can have a stand-alone device that produces the signal. But a system that requires time must compare the count of time on its local device to an external reference. This has broad implications in telecommunications systems, and in the many other systems that require some level of accurate time coordination. Not only must clocks be chosen and implemented to run properly, but their signals must be transported and measured properly.

This book describes the needs for synchronization in telecommunications networks and the current evolution of methods and standards that enable it. The challenge of supplying needed synchronization to telecommunications systems is primarily an engineering problem, not a theoretical or scientific one. Because the requisite frequency devices can be expensive, and because the necessary time synchronization must be transported, there is a need for a synchronization network. The requirements of a communications network fundamentally conflict with those of a synchronization network. The communications network ideally separates functional layers, so devices interact with other devices only one layer up or down. Synchronization requires direct access to the lowest layer, the physical layer, since synchronization, unlike data, requires a physical signal. Applications that are required to consume some form of synchronization signal can be in any layer of the communications network. Hence they must break or tunnel through the isolation of layers to get access to the synchronization signal, in violation of the layer principles of a communications network.

Alternatively, a synchronization signal can be supplied from a source external to the communications network. Receivers of Global Navigation Satellite Systems (GNSS) such as the U.S. Global Positioning System (GPS) are commonly used to provide both time and frequency synchronization. These cannot be used everywhere, however, both because of expense and because of difficulties in getting the synchronization signals where they are needed. In addition, GNSS signals are vulnerable to interference – both intentional and unintentional interference. Thus, even with GNSS signals available, a synchronization network remains essential.

As I write in 2013, synchronization in communication systems is in the midst of evolving from the role of primarily frequency synchronization to the role of precise time synchronization. The metrology community separates these two types of synchronization by calling “synchronization” in frequency the name “syntonization”, though the telecommunications community uses the word synchronization for both. The transport of networks in the late 80 s and 90 s was itself synchronous in frequency, or syntonous. With the advent of packet networks, the transport no longer needed syntonization, yet many applications and services still required various forms of either syntonization or synchronization or both.

Among other things, synchronization optimizes available bandwidth, enabling a more efficient use of the spectrum. Today, wireless networking is becoming more ubiquitous. Time synchronization is becoming essential to allow high data rates in the limited wireless spectrum. This becomes a complex engineering problem, as many different scenarios require different types of synchronization. Traditional synchronous networks still remain in use and require syntonization, while packet networks dominate all new roll-outs. Further, many size-scales of wireless networks are being deployed, from macro-cells over cities to femto-cells in a small interior of a building. These hybrid networks challenge operators to supply needed synchronization to all the requisite applications and services.

Within the context of these circumstances, this book is timely. As synchronization becomes both more complex and more necessary, there still remains a dearth of training and learning options. This book is a comprehensive effort by experts who have been developing standards and engineering devices, and employing these in real networks. It should help fill the void for those trying to negotiate the diverse and complex world of time and frequency issues in communications systems.

This book is a collaboration of major figures in the creation and use of synchronization. I will not repeat the information in the biographic, but I want to mention my appreciation and respect for this team of authors and the current effort. The authors are a mixture of standards experts, equipment building and testing experts, and operators who must implement and maintain synchronization. This book is a work of significant magnitude, involving many hours of development and coordination.

Synchronization in telecommunications is a fascinating field. It involves complex concepts and difficult engineering efforts. Concomitantly, the field creates great benefits for many users, facilitating an increasing ease for human social communications underpinned by large data transfers. In addition, synchronization facilitates large industries representing many billions of dollars. This book can take the user a long way along this river. Enjoy!

Marc WEISS Mathematical Physicist, GPS and Telecom Sync Expert at NIST

Abbreviations and Acronyms

This section lists the abbreviations and acronyms used in this book.

3GPP

Third-Generation Partnership Project

3GPP2

Third-Generation Partnership Project 2

AAA

Authentication, Authorization and Accounting

AAL1

ATM Adaptation Layer 1

ACMA

Australian Communications Authority

ACR

Adaptive Clock Recovery

ADEV

Allan Deviation

ADM

Add Drop Multiplexer

ADPCM

Adaptive Differential Pulse Code Modulation

ANSI

American National Standards Institute

ATIS

Alliance for Telecommunications Industry Solutions

ATM

Asynchronous Transfer Mode

AVB

Audio Video Bridging

BC

Boundary Clock

BIPM

International Bureau of Weights and Measures

BITS

Building Integrated Timing System

BMC

Best Master Clock

BMCA

Best Master Clock Algorithm

BNC

Bayonet Neill-Concelman (connector)

BPM

People’s Republic of China’s National Time Signal Service

BS

Base Station

BSC

Base Station Controller

BSS

Base Station Subsystem

BTS

Base Transceiver Station

CBR

Constant Bit Rate

CDMA

Code Division Multiple Access

CDR

Clock Data Recovery

CEM

(SONET/SDH) Circuit Emulation Service over MPLS (RFC5143 – obsoleted by CEP)

CEP

(SONET/SDH) Circuit Emulation over Packet (RFC 4842 – obsoletes CEM)

CERN

European Organization for Nuclear Research

CES

Circuit Emulation Service

CESoPSN

Circuit Emulation Service over Packet-Switched Network (RFC5086)

CF

Correction Field

CI

Characteristic Information (ITU-T Rec. G.805)

CLI

Command Line Interface

CNSS

Compass Navigation Satellite System

CoMP

Coordinated Multipoint

CPE

Customer Premise Equipment

CPRI

Common Public Radio Interface

CSG

Cell Site Gateway

D/A

Digital to Analog

DCF

Dispersion Compensating Fibers

DCN

Data Communications Network

DCR

Differential Clock Recovery

DCF77

Radio Time Service in Germany

DNU

Do Not Use (QL value interpretation)

DoS

Denial of Service

DPLL

Digital Phase Lock Loop

DS1

Digital Signal 1 (1.544 Mbit/s)

DSL

Digital Subscriber Line

DSLAM

Digital Subscriber Line Access Multiplexer

DTI

DOCSIS Timing Interface

DUS

Don’t Use (QL value interpretation – equivalent to DNU)

DUT

Device Under Test

DVB-T/H

Digital Video Broadcast – Terrestrial/Handheld

E1

Digital signal (2.048 Mbit/s)

E2E

End-to-End

EEC

(Synchronous) Ethernet Equipment Clock

eICIC

Enhanced ICIC (Inter-cell Interference Coordination)

eLORAN

Enhanced LORAN

EPC

Evolved Packet Core (LTE)

EPL

Ethernet Private Line

ESI

External Sync Interface

ESMC

Ethernet Synchronization Messaging Channel

ETH

Ethernet MAC Layer Network (IU-T)

ETSI

European Telecommunications Standards Institute

ETY

Ethernet PHY Layer Network (ITU-T)

FCC

Federal Communications Commission

FCS

Frame Check Sequence

FCAPS

Fault, Accounting, Configuration, Performance and Security Management

FDD

Frequency Division Duplexing

FLL

Frequency Lock Loop

FPP

Floor Packet Percentage

GAARDIAN

GNSS Availability Accuracy Reliability and Integrity Assessment for Timing and Navigation

GBAS

Ground-Based Augmentation System

GE

Gigabit Ethernet

GFP-F

Generic Framing Procedure-Framed

GLONASS

Globalnaya Navigatsionnaya Sputnikovaya Sistema

(Global Navigation Satellite System)

GM

Grand Master

GMP

Generic Mapping Procedure

GNSS

Global Navigation Satellite System

GPS

Global Positioning System

GRI

Group Repetition Interval

GSM

Global System for Mobile communications

GUI

Graphical User Interface

HL

Hop Limit

HLR

Home Location Register

HOL

Head Of Line

HOLB

Head Of Line blocking

HRM

Hypothetical Reference Model

HRX

Hypothetical Reference Connection

HSPA

High-Speed Packet Access

HSS

Home Subscriber Server

IANA

Internet Assigned Numbers Authority

ICIC

Inter-cell Interference Coordination

ID

Identifier or Identity Description

IED

Improvised Explosive Device

IEEE

Institute of Electrical and Electronics Engineers

IETF

Internet Engineering Task Force

IMA

Inverse Multiplex for ATM

IP

Internet Protocol

IP FRR

IP FastReRoute

IPDV

Inter-Packet Delay Variation or IP Packet Delay Variation

IRIG

Inter-Range Instrumentation Group

IRNSS

Indian Regional Navigational Satellite System

ITSF

International Telecom Sync Forum

ITU

International Telecommunication Union

ITU-T

International Telecommunication Union – Telecom

Iu

Interconnection point between an RNC or a BSC and a 3G Core Network

Iub

Interface between an RNC and a Node B

IWF

Interworking Function

JESS

Jitter Estimation by Statistical Study

JLOC

Jammer Location

LACP

Link Aggregation Control Protocol

LAN

Local Area Network

LBAS

Local Based Augmentation System

LORAN

Long Range Aid to Navigation

LSP

Label Switched Path

LTE

Long-Term Evolution

LTE-A

LTE Advanced

Lx

Layer x

M-CMTS

Modular Cable Modem Termination System

MAC

Media Access Control

MAFE

Maximum Averaged Frequency Error

MATIE

Maximum Averaged Time Interval Error

MB(M)S

Multicast Broadcast (Multimedia) Services

MBSFN

Multicast Broadcast Single Frequency Network

MDEV

Modified Allan Deviation

MEF

Metro Ethernet Forum

MI

Management Information

MIB

Management Information Base

MinTDEV

Minimum TDEV

MME

Mobility Management Entity

MNO

Mobile Network Operator

MPLS

Multi-Protocol Label Switching

MRTIE

Maximum Relative Time Interval Error

MS

Mobile System

MSAN

Multi-Service Access Node

MSC

Mobile Switching Center

MSF

UK Low frequency time signal and standard frequency radio station based on the NPL time scale UTC(NPL)

MTIE

Maximum Time Interval Error

MTOSI

Multi-Technology Operations System Interface

MTU

Maximum Transmission Unit

MW

Microwave

NE

Network Element

NGN

Next-Generation Network

NIST

National Institute of Standards and Technology (USA)

NMS

Network Management System

NPU

Network Processor Unit

NS

Network Synchronization (ITU-T)

NTP

Network Time Protocol

NTR

Network Timing Reference

OAM

Operations, Administration, Maintenance

OAM&P

Operations, Administration, Maintenance and Provisioning

OC

Ordinary Clock

OCXO

Oven-Controlled Crystal Oscillator

ODUk

Optical data unit of level k

OFDM

Orthogonal Frequency-Division Multiplexing

OLT

Optical Line Terminal (PON)

ONU

Optical Network Unit

OPEX

Operational Expense

OS

Operating System

OSC

Optical Supervisory Channel (OTN)

OSI

Open Systems Interconnection

OSPF

Open Shortest Path First

OSS

Support System

OSSP

Organization-Specific Slow Protocol

OTDR

Optical Time-Domain Reflectometer

OTN

Optical Transport Network

OTT

Over-The-Top

OUI

Organization Unique Identifier

P2P

Peer to Peer

PABX

Private Automatic Branch Exchange

PAR

Project Authorization Request

PCM

Pulse Code Modulation

PCP

Port Control Protocol

PDF

Probability Density Function

PDH

Plesiochronous Distribution Hierarchy

PDN-GW

Packet Data Network-GateWay

PDU

Protocol Data Unit

PDV

Packet Delay Variation

PHY

Physical

PLL

Phase-Locked Loop

PM

Packet Master

PMC

Packet Master Clock

PNT

Position, Navigation and Timing

PON

Passive Optical Network

POS

Packet over SONET (or SDH)

ppb

Part per billion

ppm

Part per million

PPS

Pulse per second

pps

Packets per second

PRC

Primary Reference Clock

PRS

Primary Reference Source

PRTC

Primary Reference Time Clock

PS

Packet Slave

PSN

Packet-Switched Network

PSTN

Public-Switched Telephone Network

PTP

Precision Time Protocol

PTSF

Packet Timing Signal Failure

PW

Pseudo-Wire

PWS

Pseudo-Wire Service

QL

Quality Level

QZSS

Quasi-Zenith Satellite System

RAIM

Receiver Autonomous Integrity Monitoring

RAN

Radio Access Network

RBAS

Regional Based Augmentation System

RF

Radio Frequency

RNC

Radio Network Controller

RNSS

Radio Navigation Satellite Service

RT

Residence Time

RTP

Real-Time Protocol

SA

Selective Availability

SAE-GW

System Architecture Evolution-GateWay

SASE

Stand Alone Synchronization Equipment

SATop

Structure-Agnostic TDM over Packet (IETF)

SBAS

Satellite-Based Augmentation System

SD

Synchronization Distribution (ITU-T)

SDCM

System for Differential Corrections and Monitoring (GLONASS)

SDH

Synchronous Digital Hierarchy

SDO

Standardization Development Organizations

SDSL

Symmetric Digital Subscriber Line

SEC

SDH Equipment Clock

SETG

Synchronous Equipment Timing Generator

SETS

Synchronous Equipment Timing Source

SFN

Single Frequency Network

SFP

Small Form Factor Pluggable

SGSN

Serving GPRS Support Node

SG15

Study Group 15 (ITU-T)

SLA

Service Level Agreement

SMA

SubMiniature Version A (connector)

SMB

SubMiniature Version B (connector)

SMC

SubMiniature Version C (connector)

SNMP

Simple Network Management Protocol

SNTP

Simple Network Time Protocol

SOF

Start Of Frame

SONET

Synchronous Optical Network

SOOC

Slave Only Ordinary Clock

SP

Service Provider

SRTS

Synchronous Residual Time Stamps

SSH

Secure SHell

SSM

Synchronization Status Message

SSU

Synchronization Supply Unit

ST3

QL Value for Stratum3

STM-N

Synchronous Transport Module (level N)

SyncE

ITU-T Synchronous Ethernet

S1

Interface between an eNB and an EPC

T-BC

Telecom-BC (boundary clock)

T-GM

Telecom-GM (grandmaster)

T-SC

Telecom-Slave Clock

T-TC

Telecom-Transparent Clock

T-TSC

Telecom-Time Slave Clock

TAI

Temps Atomique International

(International Atomic Time)

TASI

Time Assignment Speech Interpolation

TC

Transparent Clock

TCO

Total Cost of Ownership

TD-SCDMA

Time Division-Synchronous CDMA

TDD

Time Division Duplexing

TDEV

Time Deviation

TDF

TéléDiffusion de France

(radio time service broadcasted by TDF)

TDM

Time Division Multiplexing

TDMA

Time Division Multiplexing Access

TICTOC

Timing over IP Connection and Transfer of Clock (IETF WG)

TIE

Time Interval Error

TKS

Time Keeping System

TLV

Type Length Value

TMF

Telemanagement Forum

TNC

Threaded Neill-Concelman (connector)

TNM

Telecommunications Management Network

ToD

Time of Day

TSG

Timing Signal Generator

TTL

Time To Live

TTT

Timing Transparent Transcoding

TWSTFT

Two-Way Satellite Time and Frequency Transfer

TWTT

Two-Way Time Transfer (protocol)

UDP

User Datagram Protocol

UE

User Equipment

UI

Unit Interval

UMTS

Universal Mobile Telecommunications System

US

United States

USNO

US Naval Observatory

UTC

Coordinated Universal Time

UTP

Unshielded Twisted Pair (cable)

UTRAN

UMTS Transport Radio Access Network

Uu

Radio Interface between the UE and the NodeB

VCO

Voltage Controlled Oscillator

VDSL

Very High Speed Digital Subscriber Line

VLAN

Virtual LAN (Local Area Network)

VoIP

Voice over Internet Protocol

VPN

Virtual Private Network

WAAS

Wide Area Augmentation System

WAN

Wide Area Network

WCDMA

Wideband CDMA

WDM

Wavelength Division Multiplexing

WG

Working Group

WiMAX

Worldwide Interoperability for Microwave Access

WS

Work Station

WSTS

Workshop on Synchronization in Telecommunication Systems

WWWF

Radio Time Service from NIST (US)

xDSL

x (any type of) Digital Subscriber Line

Acknowledgments

Jean-Loup Ferrant

I would like to thank Alcatel, now Alcatel-Lucent, and specially Bernard Point and Bernard Sales, who supported my work on synchronization and standardization in ITU, ETSI, IEEE and IETF.

I would like to thank Tommy Cook, CEO of Calnex Solutions, who has been sponsoring my activity in ITU-T Q13 after I retired from Alcatel-Lucent.

I want to thank all the participants of ITU-T Q13 for their work during the last decade, which allowed Q13 to address the new issues raised by the transport of synchronization in packet networks.

I want also to thank my family who supported me during my work on this book.

Mike Gilson

I would like to thank BT and specifically both Tony Flavin and Glenn Whalley for their support of my work on this book and, in the wider context, their ongoing support for the synchronization subject in general. I would also like to thank the team of professionals I work with, Greg Mason, Sean Taylor and Trevor Marwick; both Sean Taylor and Trevor Marwick have worked on the evolving SyncE/1588 technology and provided me with considerable support for which I am indebted to them.

It has been a privilege to work on the development of synchronization standards and see their evolution from concepts to finished standards and then follow their adoption into deployed networks for the benefit of all. In the 1990s, during my first term in telecoms standardization, I met many inspirational people – some are still around, Dr Ghani Abbas being one of them who taught me much. During my second term from the early 2000s, I have had the pleasure to work with many new people on this subject. I feel fortunate to have met them all and worked with many, although too many to name it is primarily the collective group in ITU-T SG15Q13, ITSF and WSTS.

I would like to thank my family who have put up with the many weeks away over the years that have been due to my standards participation. Special thanks go to my partner Jan Longthorp, who has dropped me off and collected me from many different airports and who has also put up with many broken weekends resulting from both standards participation and the subsequent writing of this book. On behalf of all the authors, I would also like to thank Jan for all the help she gave us in the final proof reading.

Finally, I dedicate this book to my parents.

Sébastien Jobert

I would like to very much thank my company, France Télécom Orange, for supporting this work, and all the studies that we initiated with my colleagues in Lannion. I thank in particular Jean-Paul Cornec, my predecessor representing France Télécom in ITU-T SG15 Q13, for having kindly shared his expertise when I joined the team. I also thank Pierre-Noël Favennec for his kind help in finding the publisher of this book.

I would like to acknowledge the excellent technical work that has been done over the years in ITU-T with participants of other companies that are not part of this project (but who could have been for sure without any problem), in particular, Geoffrey Garner and Kenneth Hann (there are many others, but the list would be too long …). The 7–8 year period preceding the publishing of this book has, indeed, been very fruitful in standards and is very likely one my most exciting experiences.

I obviously thank my family very much, my lovely wife Stéphanie, daughter Jessica and son Romain, for kindly supporting the hard work and the long days behind the computer writing this book or attending standardization meetings.

Finally, I dedicate this book to my father.

Michael Mayer

First and foremost, I would like to thank my wife Zsuzsanna for her support and patience during the many months involved with the preparation of this book. Words are not enough to express my gratitude to her for accommodating my absences during the many standards meetings where much of the content of this book was honed and refined. My children Fanni, David and Ben also deserve thanks for putting up with my travels.

Much of the content of this book has been developed over the many years of collaboration with my other colleagues in various companies and standards bodies, particularly the ITU-T and COAST-SYNC. I am very thankful for the privilege to work with so many bright and talented people.

I would like to also thank my coauthors for the opportunity to participate in this truly collaborative effort.

Finally, I would like to dedicate this book to the memory of my parents.

Laurent Montini

I would like to thank the Q13/15 attendees, most coauthors, who, in 2005, welcomed the first “packet guy”, and to Marc Weiss who provided me with extended help for this book.

I would like to thank Cisco as a company for promoting thought leadership and providing me with the opportunity to meet and collaborate with so many great talents and individuals. I also thank all my managers (Axel Clauberg, Cedrik Neike, Art Feather, Jane Butler, Chip Sharp and Russ Guyrek) who trusted me in supporting this work.

I would like to particularly recognize Stewart Bryant as an early and staunch supporter and Leonid Goldin as a faithful partner on synchronization for years. I must express my profound gratitude to my very first mentor Jean Guylane, long before my tenure at Cisco.

I would like to thank my four beloved sons and daughters who have, through their patience, strongly contributed to this book. Special thanks go to my wife Marie for her love and support over the years.

I dedicate this book to my parents.

Michel Ouellette

I would like to thank my lovely wife Kim, beautiful daughters Haley and Oliva, and my parents Francois and Monique for their support and looking after me during those long evenings and weekends while I was spending time with my other wife “the Computer”.

Special thanks go to Dr James Aweya, an outstanding mentor who taught me so many things. I also thank Bob Mandeville from Iometrix for the opportunity he provided me, as well my previous colleagues at Huawei Technologies and to this great university that was once called Nortel.

Silvana Rodrigues

I would like to thank Integrated Device Technology (IDT), my coworkers, specially my former manager Jim Holbrook, and my current manager Louise Gaulin for their support of my work on this book. I would also like to thank IDT for its support of the standards activities.

It has been a pleasure to work with several colleagues from different companies at the ITU-T and IEEE standards meetings, and ITSF, ISPCS and WSTS workshops. Over the years, several colleagues became very good friends. I feel very fortunate to work with such a great group of people. I would also like to thank John Eidson who provided me with advice while writing this book.

Special thanks go to my loving family, my husband Claudio and sons Nicholas and Thomas. Their support throughout my career was fundamental for my development, with so many weeks away from home participating in standards meetings, and many weekends writing this book. I also would like to thank my sisters Celi and Ivani, and my brother Fabio for their support. I dedicate this book to my father (in memory) and to my mother.

Stefano Ruffini

I wish to thank all the people who have supported me working on this book. They include my company, Ericsson, which has given me the opportunity to take part in this project and has provided with great support over the many years in my participation in the standardization activities, and in particular my current manager, Roberto Sabella, and Ghani Abbas whose experience in the standardization activities has often been a great help; all colleagues, especially those I have met during the ITU-T Q13/15 meetings, and who during these years have inspired me and with whom I often have established extraordinary relationships; and my loving family – my son Dante, daughter Emma Vittorina and wife Elin – who have supported me in this project.

Moreover, I wish to remember the Fondazione Ugo Bordoni and, in particular, Domenico De Seta, with whom more than 20 years ago I started to learn about synchronization in SDH systems.

Finally, I would like to dedicate this book to the memory of my parents, Giovanni and Vittorina.

Introduction

I.1. The importance of synchronization in future telecommunications networks

Exchanging digital data has always required some level of synchronization: the receiver of a telecommunication system must correctly acquire the “rhythm” or frequency of the bits sent by the transmitter in order to recover the data correctly. But synchronization within a telecommunications network has a much wider scope than between a transmitter and receiver on a local link: the delivery or distribution of a common timing reference across a telecommunications network is required for various network applications to ensure proper network operation. This timing distribution effectively becomes a network within a network. In many cases, this distribution follows the same path across a network as the data. However, in some cases the data may flow across the network but the timing will flow from a central point to the edge and only follow the same path for part of the distribution, typically at the edge. Ideally engineering the synchronization network should take place at the same time as engineering the network to carry data. However, this is not always the case.

Synchronization is often thought of for time division multiplexing (TDM) based fixed line voice and data-based infrastructure networks. However, few realize the importance that synchronization plays in allowing various applications to work correctly, one such application that is gathering pace is the increase in mobile telecommunications through long-term evolution (LTE). For instance, in mobile telephony, the synchronization requirements of the air interface are critical. How would wireless mobile (based on technologies such as global system for mobile communications (GSM), code division multiple access (CDMA) and LTE) communications work without synchronization? Clearly it would not and Quality-of-Service issues would arise if it were not considered.

Quality-of-Service and synchronization have always had close links. This has been true in the TDM world where accurate synchronization is required to limit the occurrence of slips.

Synchronization is still needed today and in future for mobile applications and networks:

– to stabilize the radio frequencies used by the mobile base station;

– to allow efficient spectrum usage;

– to avoid radio interference between neighboring cells;

– to allow seamless hand over between cells.

Poor synchronization within a telecommunications network may have important impacts on the end user:

– The communication can degrade (voice communication can become inaudible).

– The throughput of data connections in the networks can reduce.

– The network’s connections (in the case of the internet) might even be totally lost.

– In the case of mobile communications, hand over between cells could fail and quality of experience degrade.

Ensuring a proper design for a synchronization network should therefore not be underestimated when considering these potential impacts.

I.2. Purpose of this book

One of the problems faced by the network engineers responsible for building suitable synchronization architectures is that synchronization is not necessarily a well-known or even a well-understood topic within telecommunications. Many engineers may have a limited knowledge about the subject and may feel uncomfortable with existing synchronization technology. Equally, there are also many engineers who have no knowledge on the subject at all and no basis on which to develop their understanding of new synchronization technology in the packet world.

Now, with the evolution toward new packet-based technology and consequently new synchronization technologies, a review of some of the key principles and concepts of synchronization and timing is useful. This is especially useful when applying synchronization to new packet-based technologies to understand how they work and where they apply, how they can be tested and what challenges may exist from a network design or operational management perspective.

Dissemination of such knowledge is critical for subject areas such as synchronization that are not common or well understood, but even more so when these new technologies are being considered in new network architectures. Like any subject area, spreading a proper understanding takes time but has wider and longer-term benefits.

Synchronization design and the vagaries around the subject is often a specific discipline that is practiced by a relatively few individuals on a day-to-day basis. These individuals tend to be experts that sit in a wide range of companies within the telecoms and associated industries. For example:

– Large network operators will often have a few individuals who understand the issues across their deployed technologies and scale of operations. When these operations cover many different types of voice and data applications and span tens or hundreds of thousands of elements, supporting many millions or even billions of dollars in revenue, it becomes apparent that the scale of this challenge can be large and the risk under failure conditions is high.

– Systems vendors will have experts in designing and integrating synchronization capabilities within their products and may well have expertise in designing these products into some aspects of the synchronization network.

– Silicon vendors will have experts in designing and integrating their components into the various systems.

– Other organizations may well have a small group of experts, for example these could be test houses or consultancies. A few specialist companies also make it their business to design specific synchronization-related products or provide expert consultancy or both.

However, it is worth knowing that the number of people worldwide with a fair knowledge in the industry is probably in the order of a few thousand (and could actually be below a thousand) and this drops to only a couple of hundred who have real knowledge of how synchronization is designed into real networks and how the industry is evolving. At events such as International Telecom Synchronization Forum (ITSF) and Workshop on Synchronization in Telecommunication Systems (WSTS) through the late 2000s, typically between 70 and 120 experts attended each year. The experts actively involved in standardization results in an even lower number: in 2003, the expert question in the International Telecommunications Union (ITU) dealing with synchronization had less than 10 experts dealing with the subject and even in 2013 at most it will be 45.

Some of the disciplines that are involved in synchronization require the engineer to:

– have a good understanding of the overall network architecture and the different technologies that make this architecture;

– understand how these technologies work (certainly at a high level) but in some cases to some depth;

– have an appreciation of digital networking;

– have a detailed understanding of analogue technology and some of the factors affecting oscillator performance and other clock components;

– have an understanding of the services carried and the performance requirements. Also how they may be degraded and what may degrade them to sort out potential synchronization problems from the normal service problems;

– have an understanding of how to test for synchronization problems in live networks and for lab-based evaluation. The engineer also needs to understand how the various pieces of test equipment work and what may influence the results, for example a badly tuned reference oscillator;

– understand the standards and what can and cannot be achieved in terms of architecture, technology and performance;

– take the network architecture that has often been developed to carry services without any thought to synchronization and work out how to add synchronization for the services that often have a demanding commercial criteria;

– have an understanding of packet networks and packet technology in the world of NGNs.

There are probably more, but hopefully the reader can see that the synchronization engineer while being a specialist in the field of synchronization also needs, to a certain degree, to be a “jack of all trades”. That is they have a wide base of knowledge that can practically be applied. In many cases, although synchronization knowledge can be acquired through study, it will often be learnt over many years of dealing with practical problems of design, problem resolution and testing.

One thing that should be clearly stated is that the content within this book is the work of many experts over many years. Much of this work has become codified within standards and has obviously been based on contributions to standards bodies such as ITU-T throughout the years, for example public switched telephone network (PSTN) synchronization in the 1980s, then synchronous digital hierarchy (SDH) technology and its respective synchronization in the 1990s. On the more recent technologies, these contributions have been from an increasing group of experts that now regularly attend ITU-T. The authors of this book have attempted to distill this collective knowledge and translate it into an up-to-date body of work. Some of this work is the authors’ own work and other aspects are based very much on the work of others distilled into a consensus view by the ITU-T process with an attempt by the authors to translate this into a useful reference source.

There may be many reasons why a reader may find this book useful:

– It brings interested engineers very much up to date with the latest technology on this topic.

– It provides some useful guidance based on the latest standardization.

– It provides a handy reference for expert engineers who need to check technical aspects on the latest technology.

This book aims to provide some clarity to the subject and highlight the importance of synchronization so that the subject is not forgotten, or considered only at a very late stage, or poorly designed inside an overall network architecture, especially in the packet network world.

As discussed earlier, experts in the subject of synchronization tend to be fairly rare, with the non-expert in the subject split between those that know the requirement for synchronization and timing exists, and those that do not. This book will have something for the non-expert readers as well and should clarify and build on their existing telecoms knowledge or provide a base on which to explore the subject further. This should help to demystify some aspects of synchronization.

Any engineer involved in synchronization will be familiar with the challenges faced when service is failing or Quality-of-Service metrics are declining and the first response is that “it must be the synchronization”. It is true synchronization is sometimes to blame. However, there are many other issues that can cause problems that look as though they are synchronization related. Certainly unfamiliarity with the topic does not help. However, the subject of synchronization essentially revolves around the simple concept of distribution of accurate frequency, time or phase from point A to point B. Some of the detail behind this is complex, but in reality it can be divided into areas that can be explained with simple clarity and intuitive examples which hopefully this book will achieve. The danger is that oversimplification misses certain aspects of the topic or creates yet further misunderstanding.

All synchronization designers have seen “The Good, The Bad and The Ugly” – to use the title from the Sergio Leone spaghetti western film starring Clint Eastwood – in terms of network synchronization designs. Sometimes bad and ugly designs are created through necessity or inherited from previous work; they may have resulted through network migration or are determined through architecture or are sometimes driven by expedient commercial needs. This book cannot comment on those reasons, but what it can do is talk about the approach taken in the development of synchronization standards, which is the key starting point in creating good reusable synchronization designs that have a solid technical foundations and are proven to interwork correctly and stand up commercially (i.e. in both capital investment terms and operation costs). Standardization provides a key coordinating framework around which equipment and their respective interfaces can be specified, network limits can be appropriately designed and how architectures can be created to meet the various performance requirements using an agreed set of design rules. This book, written by a group of people deeply involved in the standardization process of recent synchronization technologies developed over the last decade, attempts to clarify these results.

The authors have all been in various standards bodies and contributed to the development of these new synchronization technologies developed over the last decade by the ITU Telecommunication Standardization Sector (ITU-T), such as Synchronous Ethernet or Precision Time Protocol version 2 (PTPv2) telecom profiles based on IEEE Standard 1588-2008.

As indicated, one of the objectives of this book is to describe the state-of-the-art of these technologies and what can and cannot be achieved with them. It also aims to show how the standards that have been developed should be used and understood. It further discusses the evolving needs for synchronization in a 21st Century telecoms environment and illustrates some of the challenges related to synchronization and its evolution.

Another important goal of this book is to help dispel a few myths sometimes stated or answer questions sometimes asked in the telecom industry, such as:

– synchronization is only for legacy networks;

– synchronization is not needed when migrating toward Internet Protocol (IP) networks;

– Global Positioning System (GPS) is sufficient when synchronization is needed;

– Precise Time Protocol (PTP)/IEEE 1588 is the best solution for synchronization over packet networks;

– PTP is more precise than Network Time Protocol (NTP);

– why synchronize the Ethernet physical layer?

This book, as the title indicates, addresses Synchronous Ethernet and IEEE 1588 as it is used in telecom networks. This book helps to understand how and why Synchronous Ethernet technology was developed, and how it can interwork with existing SDH-based synchronization networks.

Likewise with IEEE Standard 1588-2008 (“1588v2”, or “PTPv2”), this book will also provide up-to-date information about this new technology that has been developed, when it can be applied to telecoms networks, and explain the concept of “profiles” and the different PTP telecom profiles developed or under development at the ITU-T. It will explain how to use these standards, the limitations and possible combinations. It will also show how this can be linked with Synchronous Ethernet technology when moving from simple end-to-end (E2E) frequency-based transport to very high precision time and phase transport.

The synchronization world has – like any technology area – many concepts or terms that can be used and are also misused or even used interchangeably. For example, where only frequency is concerned, the oscillators and associated filtering within equipment are collectively called a clock, but these clocks traditionally do not tell the time. However, in the context of next-generation synchronization, clocks will have some concept of a time base or transfer time. Hence timing can now also mean phase and time. Similarly, the term synchronization is often called timing but tends to mean frequency synchronization. However, again in the next-generation synchronization network, timing can now also mean phase and time. Similarly, new terminology previously not associated with traditional synchronization networks will also be introduced.

An attempt will be made to clarify some of these terms and put them into propercontext within the packet-based world these technologies will exist. For example, this book will describe the concepts of packet delay variation (PDV), a term that is often called jitter in the data world. Although both can be measured over time, the jitter in packet networks tends to represent a “maximum value” when PDV is associated to a “variation over time”, which is much more important to understand for timing recovery and can be analyzed and metrics developed to quantify. Some discussion will also take place on PDV metrics and their importance in the context of timing recovery in packet-based systems. Jitter, as a term when used in this book, is the strict definition of jitter when applied to synchronization.

Every synchronization specialist has probably felt at least once in their working career that the telecoms world is split into three groups, those who have no idea that the subject of synchronization exists, those who do have a vague idea that it exists – but know little about it, prefer to ignore or have a dangerously vague understanding of the topic and those who are involved in the subject. The synchronization world tends to be a small technical world that does not have a large body of written technical information published and books on the subject are relatively rare (see [BRE 02, SHE 09]) compared to some specializations. This book aims to help to fill the gap. Regardless of the reasons for reading this book, be it for information, for reference, to aid learning of a new subject or refreshing and enhancing knowledge on the subject and bringing oneself up to date with new concepts and technology, the authors hope that this book will be of value.

I.3. Differences between frequency and phase/time

The word synchronization has multiple meanings. Even only in telecommunication systems, it can correspond to very different notions: for example, data synchronization (e.g. synchronizing the data from a smartphone to a computer or a network, or synchronizing data between databases or between two entities of a redundant network device). These notions are all based on information that, in many respects, is a totally different notion from frequency, phase and time synchronization as used in this book. However, if you think of frequency or time as information the analogy is not so different, that is you are maintaining certain information between two points within certain limits one of the limits being time itself.

Dr Marc A. Weiss, National Institute of Standards and Technology (NIST) mathematician and GPS expert, has helpfully tried to encapsulate some of the key aspects and concepts and challenges of frequency, phase and time synchronization into the following paragraphs written during the development of this book:

   One fundamental issue is the differences among frequency, phase and time synchronization. Since a clock is a frequency device, one can have a stand-alone accurate frequency standard, such as a commercial cesium clock. Note, however that there are major differences among the frequency accuracy of different atomic clocks. Many atomic frequency standards are built for stability, not accuracy. Rubidium standards and hydrogen masers are extremely stable. Frequency accuracy is relative to the definition of the second based on the cesium atom. The most accurate devices are built and maintained in laboratories, and are accurate to better than 15 decimal places. These are not available as commercial devices. Note that chip-scale atomic clocks, while they may be based on the cesium atom, have little accuracy and stability. Their advantage is producing a signal more stable than the best Quartz oscillators with minimal size, weight and power.

   Time synchronization brings in new problems and concepts not considered for frequency synchronization (frequency synchronization is properly called syntonization). Time accuracy, unlike frequency accuracy, in principle requires transfer from a source of UTC. Time accuracy requires a method for transfer. Most commonly, this is provided from a GNSS receiver. Even with initially synchronized clocks, any two clocks will walk off from each other in time without bound. Perfectly accurate clocks will have some white noise in frequency, which will produce a random walk in time. Time synchronization accuracy means agreement with UTC, which in turn means traceability to a lab that produces a real-time estimate of UTC. In some cases, phase accuracy is the requirement instead of true accuracy to UTC. For phase accuracy, transfer among devices is still required. An additional complication is that UTC is coordinated among all the industrialized nations, and is postprocessed. Any real-time UTC signal is a prediction of the true UTC, which is available at least 1 month after any real-time signal. Thus, different sources of UTC differ in the time they are generating. The best sources disagree by about 10–20 ns.

This book will explain and use these concepts in the chapters that follow. Some simple illustrations are used to further explain certain concepts and these, together with the information contained within the book, will provide the reader with a useful reference. With the material in the book, a greater understanding of the topic will be achieved hopefully placing the paragraphs from Dr Weiss into full context. The reader is then encouraged to revisit those words.

Starting with some simple examples, Figures I.1 and I.2 show the concept of frequency, phase and time synchronization using clock faces.

Figure I.1 is an example where both clocks are synchronized in frequency. However, they have been set with a 10 min difference between them; they are running with exactly the same frequency (same time base), so in theory, they will always display a 10 min difference between them.

Figure I.2 is an example where both clocks are synchronized in frequency, phase and time. They have been set to the same time at 03:00, as they are well synchronized, they will keep on running with the same frequency and, theoretically, will always display the same time.

Figure I.1. Same frequency

Figure I.2. Same frequency, phase and time

Getting into more technical terms, this book will focus on the following types of synchronization, using the ITU definitions. Definitions are based on ITU-T Recommendations G.8260 [G.8260] and G.810 [G.810]:

– Frequency synchronization (also called “syntonization”): distribution of reference timing signals experiencing similar frequencies (within the relevant frequency accuracy and stability requirements) to a set of network nodes. However, the reference signals are not necessarily phase-aligned. Two different types of frequency synchronization are distinguished:

– Frequency-locked systems: two systems are considered frequency-locked if their frequency difference remains bounded by a given value. If frequency-locked systems are not phase-locked systems, then the maximum phase error accumulation is unbounded. Frequency-locked systems may be based on frequency-locked loops (FLL).

– Phase-locked systems: two systems are considered phase-locked if their phase difference remains constant in the long term (they might experience fluctuation in the short term). A fixed phase offset is allowed to be arbitrary and unknown. Phase-locked systems are generally based on phase-locked loops (PLL). Note that the notion of phase-locked systems is different from phase-aligned systems.

– Phase synchronization: distribution of reference timing signals whose significant events occur at the same instant (within the relevant phase accuracy requirement) to a set of network nodes. Phase synchronization includes compensation for the propagation delay of the reference signals. Phase synchronization is equivalent to the notion of “phase alignment”. Note this concept should not be confused with the concept of “phase-locking” described earlier, which is more related to frequency synchronization.

– Time synchronization: distribution of a time reference to a set of network nodes. All the associated nodes have access to information about time (in other words, each period of the reference timing signal is marked and dated) and share a common time scale and related epoch (within the relevant time accuracy requirement). Time synchronization and phase synchronization are very close concepts: time synchronization simply consists of naming the significant instants delivered with phase synchronization (e.g. with time-of-day information).

Time synchronization implies phase synchronization and phase synchronization implies frequency synchronization. In other words, if two reference timing signals are time synchronized, they are also phase synchronized, and if two reference timing signals are phase synchronized, they are also frequency synchronized.

I.4. From traditional TDM synchronization to new mobile applications

The requirement for frequency synchronization was originally developed to support the needs of early digital networks using TDM. For example, the voice network was based on the use of 64 kbit/sec switching with buffering to take account of network variations. Without synchronization and all the switches in the E2E path being synchronized, data would be clocked in and out of different switches at different rates. This would result in data loss and hence an impact on the service. Of course, buffers can be made bigger, but this results in delays on the data and, in the case of voice, it will generate echo and other voice-based impairments. Synchronizing the switches allows much smaller buffering to be used – effectively just enough to soak up timing variations. These switches would be connected together using PDH technology providing an ideal means to transparently transport synchronization at the physical layer. With the development of mobile telecommunications systems, essentially the mobile switching center nodes and base stations, in very simple terms, become an extension to this system.

With the development of SDH transmission to replace PDH transmission this also required synchronization. Essentially the SDH systems could become the synchronization transport medium that the switches were connected to. In both the PDH and SDH cases, the frequency accuracy of the source of synchronization is based on atomic standards such as cesium and is hence very high.

Concerning time, many of the early requirements were based on the need to provide a time stamp to computer-based networks for file management, database management and to indicate most recent or current file. In the telecommunications environment, time stamps were used to provide a means to time stamp alarm events and correlate alarms on different equipment for management purposes. Billing devices also required time stamps. Early on, these time stamps may have been sourced at each specific equipment in turn from low-frequency radio transmitters. As computers and packet-based networks became more prevalent, the use of packet-based protocols such as NTP prevailed. However, the requirements for many of these applications were quite crude and time stamps in the order of ± 0.5 sec only were required.

All these applications and methods to transport frequency and time are effectively now very mature. However, with the move from deterministic TDM-based networks to packet-based networks typically based on Ethernet/IP, the situation for frequency and time transfer changes. Also, the requirements at the application have changed. Mobile telecommunications are a classic example.

Mobile networks typically used PDH backhaul to connect the base station to the mobile switch site. This allowed frequency synchronization to be provided minimizing buffer slips and also provided a means to synchronize the “air interface”, that is radio side of the system. Introducing SDH did not really change this model. The use of SDH allowed much greater bandwidths and simplified the backhaul network, but still allowed the same synchronization designs to be used all primarily built around voice being the main application, with text and some data. However, with the emergence of newer more capable mobile phones and a consumer desire to use more data, the use of PDH/SDH-based backhaul no longer cost effectively provides the required data rates or backhaul/transport efficiency.

Ethernet and increasingly IP provide the new backhaul. Although this drives the cost down and in some respects simplifies the network, in other respects this provides complications and challenges. Synchronization is one such area. Packet-based networks were natively not planned for transporting or technically capable of distributing frequency synchronization. In recent years, this has driven developments such as Synchronous Ethernet and PTP to attempt to overcome these weaknesses. Other challenges have also emerged. With the consumer demand for data and hence capacity from mobile, the demand for spectrum and the efficient use of spectrum at the base station has increased. New radio frequency techniques and the desire to precisely position frequency spectrum has resulted in the emergence of new stringent frequency and phase/time requirements. These requirements allow the radio spectrum to not only be used more efficiently but also in slightly different ways.

Although the fundamental frequency requirements have essentially remained the same since the early 1980s that are traceable to a reference clock based on an atomic source, the phase/time-based requirements have now changed from the typical ± 0.5 sec requirement for alarms in the 1990s to typically the 500–1,500 ns region less than 20 years later. Both Synchronous Ethernet and PTP have been designed to deliver high-precision frequency transport and form part of the solution to allow the phase/time-based information to be transported supporting these new requirements.

I.5. Structure of the book

The book is divided into a number of chapters and a quick summary of each is provided below to allow a clear understanding of the book structure.

Chapter 1 presents an overview of the evolution of the telecom networks and o the related synchronization needs, for instance when moving from TDM to packet technologies. The increased need of synchronization in mobile networks (including phase synchronization) is also mentioned as one of the main drivers for the definition of the new synchronization technologies.

Chapter 2 introduces new technologies, such as Synchronous Ethernet and packet-based timing (in particular using IEEE 1588) that are applicable to packet networks. Basic principles of synchronization in telecom networks are summarized in order to provide some background and to put these new technologies into context. The use of Global Navigation Satellite Systems (GNSS) (e.g. GPS) and other radio techniques completes the picture in terms of synchronization techniques. Indeed, GNSS will play a fundamental role especially when it comes to distribute a common time synchronization reference.

Chapter 3 introduces some key architectural concepts to allow better understanding of the role of the synchronization networks as part of the overall telecom network architecture. In particular, the layered approach is used to better explain what the consequences are of carrying timing over packets networks. Architectural aspects for both frequency and time synchronization are introduced including some basic aspects related to redundancy and restoration schemes.

Chapter 4 introduces the evolution of TDM to packet-based networks and discusses some of the issues and considerations related to the deployment of the Synchronous Ethernet and IEEE 1588 technology within the synchronization network. In particular, this chapter addresses some important aspects that an operator (both mobile and transport) has to take into consideration when deploying these techniques, either when the mobile and transport are the same operator or in the case when one operator may be carrying timing traffic of another operator. It covers aspects of the E2E frequency transport and the impact of network stress, issues when using Synchronous Ethernet and the advantages of interconnecting into existing synchronization architectures based around SDH, and introduces aspects of time and phase transport at both the transport and access level. Some of the considerations made in this chapter are based on the experience derived from some of the network operators who actively contributed to the ITU-T work during the 2005–2008 and 2009–2012 study periods. Readers of this chapter are encouraged to look at the developing standards and, if designing networks, to carefully assess the latest standards information with the body of knowledge within this chapter and indeed this book. Operators are advised to look at their own networks and adapt the assumptions made in this chapter according to their specific network needs.

Chapter 5 discusses some of the aspects related to network management and monitoring of the synchronization network. Synchronization distribution with Synchronous Ethernet and IEEE 1588 may need consideration when planning support from network management systems. Ongoing network monitoring, specific issues for managing wireless backhaul and to the introduction of these new technologies into existing network management systems is also briefly discussed.